oolong
PulseAugur coverage of oolong — every cluster mentioning oolong across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
AI agents: Compaction vs. exhaustive sweep for large corpora
A developer has created a system that routes questions to different AI agent architectures based on the question's class. One architecture, PrimeIntellect's prime-agent, uses a 'compaction' strategy of truncating and su…
-
4B model beats Claude Opus on custom corpus, fails public benchmark
A 4-billion parameter model, oolong, demonstrated impressive performance by correctly answering a question over a 440,000-token corpus, outperforming Claude Opus on this specific task. However, when evaluated on the pub…
-
New EnSI-RAG framework boosts long-document QA accuracy
Researchers have developed EnSI-RAG, a novel framework designed to improve question answering over long documents. This system constructs an entity-centered index that separates evidence localization from answer synthes…
-
New method reveals LLM context window benchmarks are flawed
A new research paper introduces the "Distractor-Aware Truncation" method to better evaluate the true impact of long context windows in Large Language Models. The study found that naive truncation, which removes content …
-
New SWE-Pruner Pro method optimizes coding agent context by 39%
Researchers have developed SWE-Pruner Pro, a novel method for efficiently managing long contexts in coding agents. Unlike previous approaches that used separate classifiers, SWE-Pruner Pro leverages the agent's internal…